Some skin-made Vitamin D is sulfated by the liver and stored for future desulfination and use
By Claude AI July 2026
The claim traces to Stephanie Seneff (MIT senior research scientist). Her framing is that sunlight-produced vitamin D on skin is water-soluble because it's sulfated at the 3-position, travels freely in blood, and functions differently from oral D3, which is fat-soluble and needs lipoprotein packaging. She ties this to a broader hypothesis about cholesterol sulfate being co-synthesized in skin via eNOS and being deficient in heart disease and autism. That's the source Gominak is drawing on. [Seneff talk]
The sulfate metabolite itself is real and abundant. 25-hydroxyvitamin D3-3-sulfate (25OHD3-S) has been recognized as a major circulating form of vitamin D in humans since Axelson's 1985 paper. Modern LC-MS/MS work confirms it: sulfate conjugates were 18–53% of total circulating vitamin D metabolites in a serum study of 170 older men, with 25OHD3 conjugates specifically averaging 48% across supplementation groups. Standard clinical 25(OH)D assays don't hydrolyze conjugates before measurement, so lab values only reflect the unconjugated fraction. [[Axelson 1985](https://pubmed.ncbi.nlm.nih.gov/2996940/)] [Jenkinson 2022]
Where the sulfation actually happens is the liver, not the skin. Wong et al. showed sulfotransferase SULT2A1 is the responsible enzyme, and a common SNP (rs296361, ~15% minor allele frequency in whites) predicts inter-individual sulfation rates — meaning real biological variation in how much of your D pool ends up sulfated. Kurogi et al. surveyed human tissues and again found sulfation primarily in liver cytosols, with 25OHD3 and 7-dehydrocholesterol as preferred substrates. [[Wong 2018](https://pubmed.ncbi.nlm.nih.gov/29343609/)] [Kurogi 2017]
Functionally, the current best evidence is that the sulfate is a storage/reservoir form, not an active hormone. Cheng et al. showed 25(OH)D3-3-sulfate is not metabolized by CYP27B1 (the activating enzyme) or CYP24A1 (the inactivating one) — it has to be desulfated before it can become 1,25(OH)2D3. It has high affinity for vitamin D binding protein and resists renal clearance, so it circulates longer than free 25OHD3. That's a reservoir profile, not the "active water-soluble form" Seneff describes. [Cheng 2022]
Where Seneff diverges from the data is on sourcing. Her strong claim is that skin directly synthesizes vitamin D3 sulfate during UVB exposure, alongside cholesterol sulfate, via eNOS. The mainstream model has UVB producing unsulfated cholecalciferol on skin, then hepatic 25-hydroxylation, then hepatic SULT2A1 sulfation. I could not find peer-reviewed evidence for direct skin-level sulfation of D3 during synthesis; that specific claim hasn't been reproduced in the Bikle/Holick lineage.
What is on firmer ground and worth separating: sun exposure clearly produces bioactive compounds beyond D3 that oral supplements don't replicate — nitric oxide from cutaneous nitrite stores (Weller group, Edinburgh), β-endorphin, urocanic acid, α-MSH, and near-infrared photobiomodulation of mitochondrial cytochrome c oxidase. The "sun does more than D" argument is solid on those grounds even without the sulfate story, and Gominak leans on this in the transcript. [Weller review]
Practical implication: if ~half of circulating 25OHD3 is normally sulfated and acts as a slow-release DBP-bound reservoir, oral D3 should also feed into the SULT2A1 pathway once hydroxylated in the liver. Jenkinson's 2022 data support this — they saw high sulfate fractions across supplementation status categories, which cuts against a strong version of the claim that oral D3 gives fundamentally different biochemistry from sun-produced D. A head-to-head comparison of sulfate-free ratios in sun-exposed vs supplement-only individuals would be the direct test and I don't think it's been done cleanly.